This study will extend past research on college student drinking by a) examining how patterns of harassment at school (SH) and work (WH), in the context of other school and life stressors, may differentially contribute to drinking trajectories and problem drinking for women versus men college students, and b) applying a stressor-vulnerability model to examine how a variety of gender-linked risk and protective factors may influence the relationships between WH and SH and other work, school, and life stressors and trajectories of problematic alcohol use (i.e., trajectories of increasing or chronically heavy use) over time. Hypothesized moderators include relational orientation, coping strategies (emotion focused, active, and avoidant coping), perceived social support from family, school peers, and coworkers, drinking motives (tension-reduction, sociability, conformity, and power motives), same-sex drinking norms among school and work peers, and family history of alcoholism. We will also examine the extent to which the effects of WH and SH on drinking and drug use are mediated by negative affect (depression, anxiety, and anger/hostility), differentially by gender. Results of hypothesis tests incorporating these factors will contribute to the knowledge base about predictors of underage drinking, specifically the individual, school, and workplace issues that may place female students at particular risk for problematic drinking in response to stress. We propose to survey 3,000 incoming college freshman aged 18 or older from six Illinois universities and colleges (2- and 4-year; public and private; urban and rural) using a web-based questionnaire at 6 time points across four years; (1) 2 months prior to college entry (baseline), and at (2) 4 months; (3) 8 months; (4) 20 months; (5) 32 months; and (6) 44 months after college entry. This design will allow us to explore our hypotheses using both traditional variable-centered statistical approaches to look at structural relationships among variables, as well as a person-centered approach to examine predictors and outcomes of distinct patterns of intra-individual change in drinking among women and men college students. The person-centered approach will also allow us to examine at which time points predictive factors are most salient. This will be the first study to include WH and SH as potential predictors of alcohol use in the context of other school, work, and life stressors, and to incorporate these measures into person-centered latent growth mixture models of drinking trajectories. The long term goals of this study are to: 1) better understand how the complex array of psychosocial risk factors for alcohol abuse, within a variety of domains, differ over time for young adult women and men college students, 2) use this knowledge to develop timely, gender-specific prevention and intervention strategies that might be implemented at the individual, school, or workplace levels to address alcohol abuse among college students, and 3) provide data to support legal efforts to prohibit generalized workplace harassment.